[04:31:42] tgr: order shouldn't matter for array_merge_recursive [04:36:11] oh, right [04:36:22] but for array_replace_recursive it does [04:38:29] right [04:38:36] I don't understand why that one was added [04:38:55] but I haven't gotten around to investigating/deprecating it [04:45:23] it's only used by JsonConfig iirc [14:00:23] CindyCicaleseWMF: I see you're changing around the team workboards. Where should I put reactive tasks, that I would formerly put on "Core Platform Team (CPT-Q1-Jul-Sep-2018)"? [14:28:08] anomie: this is the new workboard for reactive tasks: https://phabricator.wikimedia.org/project/view/3637/ [14:31:09] Thanks [14:32:24] CindyCicaleseWMF: still no invite, not in the calendar and not in the mailbox. [14:32:28] this seems to be stuck in fubar land [14:32:50] oh, and also: new nick ;) [14:32:57] dues: weird. [14:33:14] dues: like the new nick - shorter to type ;-) [14:33:44] nobody types nicks, is your friend [14:33:48] * anomie types "du" now, instead of "da" [14:33:58] exactly [14:43:21] dues: not duesWMF? :P [14:52:23] :P [14:52:38] He was DanielK_WMF at one point. Then I think he decided that since almost no one else uses "WMF" as a suffix on IRC he'd not do so either. [14:53:13] (for about 5.5 minutes, it looks like) [14:56:46] Clearly, I really need a new nick ;-) [14:57:04] OMG, how did I not know about for nicks? [14:57:35] dues: It says that you declined the meeting again, so it must somehow be autodeclining? [14:57:37] * addshore facepalms :P [14:58:33] CindyCicaleseWMF: I only found out last week about this little edit button on phab https://usercontent.irccloud-cdn.com/file/b0quNDAL/image.png [14:58:58] somehow I have chosen to totally ignore it for the past years [15:00:04] addshore: I think that link is actually relatively new (last couple of Phab upgrades new at least) [15:00:19] addshore: Oooh, I never noticed THAT either! I've learned so much this Monday morning already that my head is going to explode. [15:00:37] bd808: ooooh, maybe that explains it, I just saw Lydia do it last week and was so shocked I had been missing it [15:00:41] eh? I think that's old, like over a year [15:01:09] it only shows up when you hover over it, though, so it is easy to miss. That's my excuse, and I'm sticking to it. [15:01:17] greg-g: fair enough, but at least not a forever feature :) [15:01:25] * addshore likes that excuse and wil also use it [15:30:43] CindyCicaleseWMF: can you give me a link to the hangout? [15:30:47] I still see nothing in my calendar [15:31:19] dues: https://meet.google.com/xvc-gffj-ije [15:31:45] thanks [15:32:15] gah, why is google changing my accounts around all the time?! [16:24:03] dues: Are you working on T189220 (specifically, AbuseFilter and SpamBlacklist) or can I pick that up? [16:24:04] T189220: Ensure that relevant extensions are MCR-aware - https://phabricator.wikimedia.org/T189220 [16:24:36] I'm glad we don't have to make sure EP is MCR-aware [16:24:42] * James_F grins. [16:25:01] Reedy: Migrating TMH to use MCR for subtitles is going to be *so* fun, though. [16:25:03] nope the fsck out of that one [16:32:39] James_F: isn't that blocked on having the finalized PageUpdater interface, and on SlotRoleHandler/PageTypeHandler landing? [16:33:39] ...I guess in the case of readonly extensions, not really [17:15:49] tgr: It is? Should I mark it as blocked on T194046 or T196087 (and should those two be linked)? [17:15:50] T196087: Refactored implementation of MCR page update interface - https://phabricator.wikimedia.org/T196087 [17:15:51] T194046: Introduce SlotRoleHandler and SlotRoleRegistry for declaring slot roles. - https://phabricator.wikimedia.org/T194046 [17:19:10] James_F: as long as the extension does not try to write to non-main slots, just read them, I guess those changes are not really relevant. PageUpdater definitely not. SlotRoleHandler will add human-readable slot names which probably should be used in error messages, but that's a trivial detail that can be filled in later. [17:21:45] IIRC AbuseFilter uses PreparedEdit which does not expose per-slot info currently, so maybe that will be a problem? [17:23:57] Sounds like it will. [17:24:56] But yes, specifically we need to stop abuse/spam being added to file captions, so reacting to edits in non-main-slot locations. [17:25:27] We don't necessarily need to know "where" the problematic content is, just be able to return false or whatever in the hooks. [17:25:50] I can also drop it for a few days if it's too premature, which it sounds like it is. :-) [17:57:08] >This is on MediaWiki 1.31, built using the official MediaWiki container on CoreOS. [17:57:14] We have an official mw container? [18:00:33] https://hub.docker.com/_/mediawiki/ ? [18:05:01] Reedy: for certain values of 'official', yes. I don't think anyone gets to maintain it as their day job, but https://phabricator.wikimedia.org/project/profile/3094/ [18:05:13] heh [18:18:39] RoanKattouw: for when you have a minute: https://gerrit.wikimedia.org/r/c/mediawiki/extensions/FlaggedRevs/+/463816 [19:45:07] CindyCicaleseWMF: is the kanban board replacing the Q1 board? [19:48:37] legoktm: yes. Q1 is over, and rather than creating a new milestone board each quarter, I'm going to give this a try. I modeled it after what some other teams are doing, including release engineering (https://phabricator.wikimedia.org/project/board/20/) [19:49:15] alright [19:49:16] there will be a separate project/workboard to bin tasks into categories. [19:49:53] CindyCicaleseWMF: don't take us as models to copy ;) We're not the best. I actually had it on my mental backlog to look closer to how you were doing things :P [19:52:06] heh [19:52:23] greg-g: lol - oh well. I'm still in the middle of switching, so there will actually be two separate boards: one to represent status and the other two represent bins of related tasks for programs so we'll be able to visualize both aspects. We'll see how it works out . . . [19:52:43] bins being? [19:52:45] the other "to" represent . . . [19:52:47] like programs or? [19:53:01] sorry, big P Programs. [19:55:40] Yes, Programs, although some Programs might have several columns to help break things down. For example, for MCR, which will fall under the Platform Evolution Program for now, we're thinking: MCR Tech Debt, MCR New Features, MCR Reactive, MCR Uncategorized. [19:56:45] So, you'll be able to look at a task and see both that it is CPT MCT Tech Debt and CPT Kanban In Progress. [19:56:52] MCT=> MCR [19:58:23] ah, right on [21:25:05] doesn't "kanban" just mean "board"? [21:25:13] so it's a "board board"? [21:25:21] welcome to english :) [21:25:52] *sigh* [21:30:33] dues: https://en.wikipedia.org/wiki/Torpenhow_Hill :-) [21:31:56] dues: it would be easier to accept your signs if you had stoped the wikipedia/wikimedia/mediawiki situation back in the day. :) [21:32:03] *sighs [21:32:55] i was a couple of years too late for that, sorry [21:33:45] bd808: i did oppose the suggestion to use "datawiki" as the name for the extension that drives wikidata though :P [21:34:37] that would probably be even worse than wikibase :) [21:46:23] All your data, are belong to everyone. [21:52:40] bd808: yea, i wasn't a fan of "wikibase" either. but i don't even remember my own proposal now :/ [21:54:14] dues: snaktable? [21:54:46] heh [21:55:26] "Snak" was actually my idea. I recently told the story again. I was just hungry, and making a bad joke. it stuck. Yay. [21:57:20] Maybe we should rename it to Snek. But that's too 2015. [23:34:28] Reedy: Do we have a convenient way to find methods with a @deprecated but no wfDeprecated() call? [23:35:14] Not really :( [23:35:24] Sounds like the sorta thing phpcs could be made to do [23:35:38] Perhaps. [23:36:09] I searched for `// wfDeprecate` and "fixed" the only one in core last month. [23:37:28] Filed as T205938 [23:37:29] T205938: Warn when a method/class has a @deprecated annotation but no call to wfDeprecated - https://phabricator.wikimedia.org/T205938